Converting the AD797 MC Phono Pre to a MM Phono Pre - A Mixed Design.

I've got my ARC SP3-A1 on my bench.
I've got a low noise oscillator from my HP339A,
I can set at 1kHz at 1Vrms to start.
I'll fee that into the phono circuit of the SP3-A1
I'll take the phono output and run it through
my distortion analyzer. And notate the result.

I can try a couple of different test frequencies
lower, higher. Measure them.

At least I'll have a baseline from which to compare
to when my boards arrive and I build them (longer time frame)
compare them to your known measurements of your AD797 phono (short time frame).

Other's might be interested in following along and it would be measuring
prior to the board I make.

Later when the board is finished, I'll do the same tests
using the same equipment. I can also take pics of the
set up etc.
I hope that it's not really your intention to apply 1vrms at 1kHz into the preamp phono input (that would be bad...). The input should be 5.5mv rms at 1kHz and conform to the RIAA preemphasis curve if you deviate from 1kHz. The output from the new phono stage is intended to be c.1vrms.
Theoretically, the distortion should be more or less unmeasurable using the new design. However, it might be interesting to explore the overload characteristics.
 
Last edited:
I think I did. You did see my V(onoise) plot in post #227, yes? maybe refresh browser then
check.
That does not look like what I would expect.
noise1.PNGnoise2.PNG
The first is the unweighted noise for the 5.5mv rms sensitivity preamp.
The second is the A-weighted noise.
 
1. Pre amp on hold. No, I haven't done anything with it yet.
Completly untested.

2. Let me find the last unmodified version of the schematic and I'll start from there.

3. what does the c. mean when you say c. 1vrms?

4. This v(onoise) plot is different than yours.
 
Last edited:
Starting over again.
Here is my first v(onoise) plot:
01.Post245 RestartDataFixed.jpg02.RIAA restart Output.jpg
Here is the noise data for this plot:
01.POST245 RestartData.jpg

Confirmed, they are the same.
 
Last edited:
Not sure about anybody else but I'm still lurking in the back ground.
When I started my intention was (Still is) to build a better phono stage.
I have wanted to get somewhat acquainted with LT spice time permitting.
I'm not seeing the free time being available anytime soon but hope springs eternal.
If so I'm sure thread will prove a valuable reference and I thank both of you for continuing it.
 
Thanks NJ. It is always good to know that someone is following along.
I'd like to do a series of measurements to confirm what I already have
in my pre amp phono stages.
Audio Research SP3-A1
Audio Research SP-10MKII
Marantz 3250

Then when I do the boards:
MM LME49710 (from AD797)
MC AD797 Board

I'll have a starting point with baseline data for each individual
stage. At some point whole system data too.
When I put something in the system, what ever I pull out
can go to the bench for a Measurement.

It's good to know and it should be of benefit those following
along or those who are in need of baseline info for their units.
 
I hope that it's not really your intention to apply 1vrms at 1kHz into the preamp phono input (that would be bad...). The input should be 5.5mv rms at 1kHz and conform to the RIAA preemphasis curve if you deviate from 1kHz. The output from the new phono stage is intended to be c.1vrms.
Theoretically, the distortion should be more or less unmeasurable using the new design. However, it might be interesting to explore the overload characteristics.

Yes, it is a phono section...my bad.
5.5mv rms Input.
We'll see what the output looks like at 1vrms output :)
What would be a safe reasonable overload voltage NOT 1vrms?
There must be some general rules for testing of phone stages, yes?
Anyone know?

Google is our friend. These look promising from Hi-Fi World:
Phono Stage Tests: LINK
Gain
Overload
Noise
All Pages

Time to read what the site has to offer.

BTW anyone offering free tickets and passage
from Tejas to the Birstol Hi-Fi Show
February 22 thru Feburary 24 2019?
 
Last edited:
Yes, it is a phono section...my bad.
5.5mv rms Input.
We'll see what the output looks like at 1vrms output :)
What would be a safe reasonable overload voltage NOT 1vrms?
There must be some general rules for testing of phone stages, yes?
Anyone know?
At 1kHz the output stage clips at about +17dBv. This is about 15mv rms at the input or about 22mv peak. Why go beyond this at 1k.
 
That is the maximum range I should monitor.
I should start 0mv /shorted input for Noise floor.
Then stepped 1mv, 3mv, 5mv, 5.5mv for standard MM cart.
then for overload:
6mv, 7mv, 9mv, to onset of distortion...
estimate to happen around15mv.

I'll keep my eyes on the scope and when the signal begins to flatten....
document that point. I posted a link to Hi-Fi world that covers this topic.
 
Great, I see now HIFIWorld (HFW) starts with white noise.
Through an accurate inverse RIAA network.

That is for frequency response,
measured using a manual swept sine wave generator?
If manual, then each frequency's output has to be adjusted.
And we do have the Reproducing Characteristics B.S. 1928:1965
table of values. :)

GAIN

Benchmark:
MM x100 gain = 40dB
MC x1000 gain= 60dB

Ideal:
MM x200gain 46dB
MC x1k-3kgain 60dB - 70dB

Well, HFW measures theirs phono sections at the following input levels:
MM 10mV
MC 1mV

Ideal (my observation)
10mV is roughly twice the standard MM cartridge output rating of 5.5mV.
So they are in effect measuring and amps overload characteristic above the cartridges
standard output value...if I understand what the article is saying. I don't know this is
my first time. So if I'm open for suggestions to cut measurement time.

My question about "How they measure it"
Gain at 1kHz nput voltage is measure direct (i.e. p.d and not generator emf).?

I have no clue what "p.d" vs. "generator emf"?

Here is a link from the national archives, it is the
Tektronix Cookbook of Standard Audio Tests: LINK

Having this reference will at least give us a common vocabulary
and describe general test procedures, test setups, load matching
etc. Is it dated, yes. These tests should be able to be made by
anyone who is a semi-serious hobbiest.
 
Last edited:
Great, I see now HIFIWorld (HFW) starts with white noise.
Through an accurate inverse RIAA network.

That is for frequency response,
measured using a manual swept sine wave generator?
If manual, then each frequency's output has to be adjusted.
And we do have the Reproducing Characteristics B.S. 1928:1965
table of values. :)

GAIN

Benchmark:
MM x100 gain = 40dB
MC x1000 gain= 60dB

Ideal:
MM x200gain 46dB
MC x1k-3kgain 60dB - 70dB

Well, HFW measures theirs phono sections at the following input levels:
MM 10mV
MC 1mV

Ideal (my observation)
10mV is roughly twice the standard MM cartridge output rating of 5.5mV.
So they are in effect measuring and amps overload characteristic above the cartridges
standard output value...if I understand what the article is saying. I don't know this is
my first time. So if I'm open for suggestions to cut measurement time.

My question about "How they measure it"
Gain at 1kHz nput voltage is measure direct (i.e. p.d and not generator emf).?

I have no clue what "p.d" vs. "generator emf"?

Here is a link from the national archives, it is the
Tektronix Cookbook of Standard Audio Tests: LINK

Having this reference will at least give us a common vocabulary
and describe general test procedures, test setups, load matching
etc. Is it dated, yes. These tests should be able to be made by
anyone who is a semi-serious hobbiest.
The "normal" MC and MM output levels are 0.25mV and 2.5mV, respectively, measured in the usual way- 1kHz, 5cm/sec. Generally, the max level that is recorded on an LP, and the point at which the cartridge must maintain tracking, is 20cm/sec- or 4x. This leads to a max input level of 1mv and 10mv- hence the test conditions.
Your signal generator most likely has a 50 ohm output impedance and in my case I "pad down (p.d.) " the output by putting it through a resistive attenuator to reduce the level and also the source impedance- for example loading a 50 ohm generator into 5.6 ohms drops the level to c.1/10 of the unloaded (generator) output level and drops the output impedance to 5.0 ohms. This also improves the effective resolution of the output by a factor of 10 which might also be helpful. You can increase the attenuation, reduce the output impedance, and improve the resolution by padding down even more.
In order for the distortion measurements to be meaningful you need to determine and remove the source distortion levels from the measurement. There are various ways to do this.
 
Last edited:
That makes sense.
I have a couple of choices here, The HP339A is very accurate and has clean sine waves
with minimal distortion, <= -120dB. It is 600ohm.

I redid my Leader LFG1300S which is only .05% distortion at 50 ohm output.
To pad down (p.d.) the output, I can feed it into my HP350D attenuator.
As you mentioned, I can drop that 50ohm output impedance even lower.
To do that, I'd have to include something in parallel with the output to bring
the impedance down to 5.0 ohms.

I'd have to figure this out as I've not done this before. It will take me a bit
to find and calculate what I need. which will take a bit.

Once I have the process down and documented the next unit should just
be a simple job to replace the DUT (Device Under Test) interfacing to
what is already in place, right? We know there won't be any surprises
and please, no one invoke the "M" work or correlarrys (sp).

The file is too large to upload. :(

ARC SP3-A1 Specs FrontSM.jpg ARC SP3-A1 Specs RearSM.jpg

From what I think I understand now, at least with some vintage
gear, they generally weren't designed to accommodate a MC cartridge.
Max Input = 300 mv rms @ 1kHz (Magnetic phono input)
high level inputs overload proof. (unless I forget and swamp
it with 1V rms) NOT.

The Frequency Response has there are two specs preamp and phono section.
10Hz to 30 kHz, IHF load
30Hz to 30kHz, RIAA.

Here we've been discussing the RIAA curve and designing the modded
AD797 MC PRE for the LME49710HA MM PRE in a mixed configuration.

The two gain specs;
57dB magnetic phono input, main outputs IHF load.
34dB magnetic phono input, tape outputs IHF load.

Input Impedance:
Magnetic phono 50k ohms typical (150pf shunt)
High Level; 100K ohms typical.

Looking at the schematic, I should see the 150pf shunt capacitor, also.

Here is phono stage and the preamp.
01.01 Upper Left SP3-A1.jpg 01.02 Upper Right SP3-A1.jpg

Here is the power supply:

01.03 Lower Left SP3-A1.jpg 01.04 Lower Right SP3-A1.jpg

For the power supply board, I'm measuring it
to fit with individual shunts for all three output
sections of the secondary, along with upgrading
the bridge with modern diodes. The 1n400x diodes
have just about the worst specs of diodes.

For the diode bridges in the preamp I'll go with the
VS-ETH0806 (8A, 600V) FRED pt $0.86.
Ranked from best in class, citation below.

Johnson, M. (2015). Soft Recovery Diodes Lower Transformer Ringing by 10 - 20X.
Linear Audio, Volume 10. Linear Audio of Turnhout, Belgium.

Individual article available for purchase here: LINK
 
Last edited:
Have now a nice selection of parts for the boards.
What I don't know,m Wyn, which are the noiseless resistors
that you favor etc? Are we talking the bulk foil types here?
Or, some of the tantalum flavored resistors?
Or, a surprise?

In the mean time, got the internet up and running again.
No boards yet.
Ran into some issues with the preamp on the bench
along with life and then school starting up again.

Lot's of good stuff going on...stay tuned.

Another phono question, going through some references,
Any one know what happened to the ceramic cartridge?
 
Have now a nice selection of parts for the boards.
What I don't know,m Wyn, which are the noiseless resistors
that you favor etc? Are we talking the bulk foil types here?
Or, some of the tantalum flavored resistors?
Or, a surprise?

In the mean time, got the internet up and running again.
No boards yet.
Ran into some issues with the preamp on the bench
along with life and then school starting up again.

Lot's of good stuff going on...stay tuned.

Another phono question, going through some references,
Any one know what happened to the ceramic cartridge?
There is no such thing as a noiseless resistor by definition (thermodynamics, quantum mechanics at work). Metal film 1% resistors are fine.
Noiseless resistors appear in the schematic simulations as a tool to allow noise simulations that are independent of the measurement or source and it's just something that you can do in spice. I have no particular bias towards boutique resistors. As far as I know generally available metal film resistors have unmeasurable distortion and are as close to being an inherently linear element as exists.
 
That is good to know Wyn. I was wondering about that when I viewed the schematic.
Noiseless resistors. Thinking to myself, that doesn't sound too agnostic to me.

Since I don't have boards yet, I don't know what the spacing is like for the bigger caps,
i.e., the 10uf x 100V Panasonic ECQ films, I bought out the last stock they had which was
only 26. They aren't too large physically but then I'm not so sure
about the board lead spacing. I do have some fairly large Teflon caps and trying to put those on
the board probably isn't going to work. I could put them on the side and run some lead wire to them.

These are just free association thoughts here nothing cast in stone.
There is a saying for us, Cast in Stone. That would have to be lava
hot to make a casting.

Metal Films resistors do have barely measurable distortion levels, the surprise came
when an acquaintance didn't have any in the correct value needed so just stuck in some
of the standard carbon film resistors...when he measured them, the distortion also was
on the verge of unmeasure-ability. So low he stated that it shouldn't even be noticeable.

I have another LTSpice question, from the schematics that you've provided which have
signals attached to them. Are those values suitable for use with other simulations?
That is, using them for a standard preamp or a power amplifier?

Finally, can LTSpike import a schematic bmp or jpg and make a working
schematic that can be used for simulating?
 
That is good to know Wyn. I was wondering about that when I viewed the schematic.
Noiseless resistors. Thinking to myself, that doesn't sound too agnostic to me.

Since I don't have boards yet, I don't know what the spacing is like for the bigger caps,
i.e., the 10uf x 100V Panasonic ECQ films, I bought out the last stock they had which was
only 26. They aren't too large physically but then I'm not so sure
about the board lead spacing. I do have some fairly large Teflon caps and trying to put those on
the board probably isn't going to work. I could put them on the side and run some lead wire to them.

These are just free association thoughts here nothing cast in stone.
There is a saying for us, Cast in Stone. That would have to be lava
hot to make a casting.

Metal Films resistors do have barely measurable distortion levels, the surprise came
when an acquaintance didn't have any in the correct value needed so just stuck in some
of the standard carbon film resistors...when he measured them, the distortion also was
on the verge of unmeasure-ability. So low he stated that it shouldn't even be noticeable.

I have another LTSpice question, from the schematics that you've provided which have
signals attached to them. Are those values suitable for use with other simulations?
That is, using them for a standard preamp or a power amplifier?

Finally, can LTSpike import a schematic bmp or jpg and make a working
schematic that can be used for simulating?
Carbon film resistors are noisy. I would tend to avoid them for this and other reasons.
You can use any signals in any way you wish. They do have properties that should be kept/used. Click on the symbol to see the properties.
I know no way of importing a drawn schematic. Sorry.
 
Yea, the boards came in.
They seem okay, maybe a tad flexible.
I really don't like to see a trace that that
is marked for voltage going adjacent to
and tangent with the standoff hole. On the IN L side.
The trace starts at the V via, connects to the 2K gainL pot,
runs along the outside around the 1K ImpL pot
curves around the standoff then outside the plus via
and feeds the AD797 chip.

I obviously bought the wrong caps 10uf 100V.
are sized like the 2.2uf cap at the output.

I'll have to go through the board and
reconcile the difference between the parts
called out on the board and those you have provided
to me with your schematics Wyn.

Anything I should be aware of while I undertake this task?
 
Yea, the boards came in.
They seem okay, maybe a tad flexible.
I really don't like to see a trace that that
is marked for voltage going adjacent to
and tangent with the standoff hole. On the IN L side.
The trace starts at the V via, connects to the 2K gainL pot,
runs along the outside around the 1K ImpL pot
curves around the standoff then outside the plus via
and feeds the AD797 chip.

I obviously bought the wrong caps 10uf 100V.
are sized like the 2.2uf cap at the output.

I'll have to go through the board and
reconcile the difference between the parts
called out on the board and those you have provided
to me with your schematics Wyn.

Anything I should be aware of while I undertake this task?
The supply routing should be of no concern. Don't forget the opamps have enormous PSRR in the audio band, and as long as the PSU is decent there should be no problem.
There are no issues that I am aware of that you need to keep in mind. Perhaps Sachin or hypnotoad are aware of something.
 
Thanks Wyn, that is good to know.
Now all I have to do is find the original schematic
to identify the designated part number for each location
on the board, then ID your changes in that position.

Easier said then done.
 
Back
Top Bottom